WPCn ^MTv~VFa*^_P,ڨu-O&K&Ax 0SʫcѸQ,rW K$ n t6++U'I8<]GY)ތ .(9?$ D=zeР-^*vˡF~VŤ&he$XҬ-^r[7e(.߽+CD+-.HV[rI/|p mm(8W. k ɺ_e8JݤJ]' CS N8?Zx\ĦxL#QC7B邶aQvdMe.ãY5}( l~~baSr[ 4`я95I5$7"/Io=-KGF2&mרa.Y59sbZaE%{s Qg2V|ud & !+UN %\ 0b L D^ , { { F 0N: T;^ {tU { {5U< {2q {:S {:= 0Dw {[ 0mL 0e 04"#~$w@6%Lv%4%% 0% B&.& {'(:** 0[+ 0- AM../ 0D50 D+y0U F0 {0 {1:J22:]8 88 0g9UF: {9;:; {/<:< {%= AQ=:&>`> 0D&? D/j? B?o?UF%B {kBACUF\E:EUFE"FUN H DCZH B-HUFHIUFIIUFAJAJAJUFJ AJjK D3M 0KBM AYMM 0FNUNNUF@OOUNvQUFQ { R 0 R:SSSSSUNSSUN6T6T6T6T6Toe;TTo- 0 0 0z? 0r 0^+ 0 AM!(hH  Z 6Times New Roman RegularX(;$( ` TT     c:\office\wpwin\template\memofile.wptc:\office\wpwin\template\memo.wpt " i%$TT 4   2     XXVS  Theproblemdoesnotonlyconcernagriculturebutalsoanimal  husbandry,forestry,andfisheries.Inanindirectway,itmayalsoaffectotherindustries,especiallyindustriesprocessingperishablenaturalproducts.However,forsimplicityreasons,thispaperwillfocusonagriculture.# VS##X}X  k#($( ` TT    "  ($( ` TT      (3d$ ;!( ` TT       0@BB(  1  ڀ  1@BB(  1  ڀ('$( ` TT        j@BB(  1  ڀ@IIIProsandConsofAttribution(g$( ` TT      (;$( ` TT  :   @BB(  1  ڀ@ IIIProsandConsofAttributiontoNonharvestQuarters\ `$Times NewRoman(ç$( ` TT    A    i%$TT 4   4     XXXS  Seeparagraph10.107thatstates Inanycase,provisionalestimates  ofadditionstoworkinprogresscanalwaysbecalculatedonthebasisofcostsplusamarkupandrevisedwhenthevalueoftheharvestedcropbecomesknown.Seealsoparagraph6.96thatrefersto#XXXXSk#XSXXXexamplesof X intermediateconsumptionandcompensationofemployees.# XS##X}X  # ( i%$TT 4   7    gXXXXԀgSXgXThesalesbyrecipientsoutofincomeinkindmaybetoother  households,butmayalsoinvolvewholesaleandretailtradeasanactivity.Then,trademarginsshouldalsobeestimated.However,asthisisnotaspecificproblemforquarterlyaccounts,thiswillnotbefurtherdiscussedhere.# gSk##X}X  #(;$( ` TT     0@BBB  1  ڀ(;$( ` TT     0@BBB  1  ڀ(#$  0  B*MemoMemo Format  TRY 3'Memo3' 3' Letter3'T)  xH4x(  !#%t'XHP <4LegalLandLegal Landscape Orientation with Portrait NumberingTRZ'ALegal Landscape3''ALegal Landscape3'T  p0:*&hx < ` E=X X =X p B@ ,@    1  A  ԀH 6LetterLandLandscaped Orientation with Portrait NumberingTRP'3 Letter Landscape3' '3 Letter Landscape3'T  p0$:*&hx < ` Et/X X t/X pjTle4LegalPortLegal Portrait OrientationTR[A' Legal3'A' Legal3'TсLevel 1Level 2Level 3Level 4Level 5 t '5'TT  @)  1  ڀ .P/13P4D5P68D9U: B@ ,@    1  A  ԀN"vHM"3|x:(3d$ ;!( ` TT       d $ 'd'TT  U,XXXo[X` hp x (#%'0*,.8135@8:<H?AXo@BBB  2  ڀ X (;$( ` TT  ,    i%$TT 4   6     XXdS  Thiswouldbedifferentifdairyandcattlefarmingwouldbetaken  intoconsideration,becauselivestockistoberegardedasfixedcapitalassetsifitisraisedforbreedingorfurtherproduction(e.g.,milk,wool).# dS##X}X  k# 0@BBB  1  ڀ̄Level 1Level 2Level 3Level 4Level 5B*MemoMemo Format  TRY 3'Memo3' 3' Letter3'T1  xH4x(  !#%t'XHP <(3d$ ;!( ` TT      -/23A<< C $ 'd'TT  U,XXXo[X` hp x (#%'0*,.8135@8:<H?AXo@BB(  3  ڀ X ($$   1  'dxd<09Z .Courier New Regular(;$( ` TT  9  (;$( ` TT  7   0@BBB  1  ڀ  i%$TT 4   3     XXTS  ԀInthe1993SNA,severalstatementsaremadetoidentify  catastrophicevents,buttheoutcomeofthesestatementsisnotentirelyclear.Ontheonehandparagraph12.36seemstorestrictcatastrophiceventstosingulareventsofageneralnature.Examplesgiveninclude majorearthquakes,volcaniceruptions,tidalwaves,exceptionallyseverehurricanes,drought,andothernaturaldisasters.Paragraph12.37supportsthisasitrefersto abnormalflooding.Ontheotherhand,thelatterparagraphalsoincludesdroughts,withoutfurtherqualification.Itisnotparticularlyclearwhythelattershouldalwaysbedeemedcatastrophicevents,andfloodingonlyifitisabnormal.Toachieveconsistency,ageneralcriterionshouldbeused,anditisrecommendedheretolimitcatastrophiceventstoabnormal,unfrequentoccurrences.#TXXTSk#ԀTSXTXThis ]   recommendationissupportedbythediscussioninthe1993SNAoflossesof M  fixedassetsaspresentedinparagraph6.178,whichalsousesisthefrequencyoftheoccurrencesasthecriterion#TXXTS7#TSXTX.# TS##X}X  i# 0@BBB  1  ڀ Level 1Level 2Level 3Level 4Level 58T76LetterPortTRY 3' Letter3' 3' Letter3'T.Courier New Regular($( ` TT        0@BB(  1  ڀ(g$( ` TT    ?   0@BB(  1  ڀ(;$( ` TT  D  E /FGC<< C 0@BBB  1  ڀ̃Level 1Level 2Level 3Level 4Level 5($$   1  HH' dxdP Pd e i%$TT 4   5     XX[S  Thismakestherecommendationofparagraph6.96ofthe1993SNAto  reallocateoutputfor#[XX[Sk#Ԁ[SX[Xunincorporated#[XX[S9#Ԁ[SX[Xenterprises inproportiontothe  unpaidhoursworkedbytheowner(s)somewhatimpractical.#[XX[S##X}XX[X#<9Z .Courier New Regular(;$( ` TT  N   7 i%$TT 4   1     4XXXX4SX4XPresentedby# 4Sw#4S  AdriaanM.BloemU C # 4S#4S  ,ChiefoftheRealSector !    4  Division,StatisticsDepartmentoftheIMF.# 4SS##X}X  #<9Z .Courier New Regular 0@FFF  1  ڀ<9Z .Courier New Regular  i%$TT 4   2    Q XXTheproblemdoesnotonlyconcernagriculturebutalsoanimalhusbandry,forestry,andfisheries.In  anindirectway,itmayalsoaffectotherindustries,especiallyindustriesprocessingperishablenaturalproducts.However,forsimplicityreasons,thispaperwillfocusonagriculture.#QX}X  k#(hH  Z 6Times New Roman RegularSSSS7Hairlinedxd)Hairline d<9Z .Courier New Regular i%$TT 4   1    ݀PresentedbyTheauthorisChiefoftheRealSectorDivision,StatisticsDepartment  oftheIMF.<9Z .Courier New Regular<9Z .Courier New Regular<09Z .Courier New Regular!/_3 I&mage <=8C HKKKK  i%$TT 4   6    c XXThiswouldbedifferentifdairyandcattlefarmingwouldbetakenintoconsideration,because  livestockistoberegardedasfixedcapitalassetsifitisraisedforbreedingorfurtherproduction(e.g.,milk,wool).#cX}X  k#Ԁ(((('dxd( $ Figure  1  a/b3M << deUULevel 1Level 2Level 3Level 4Level 5(n$ (  1  ) (hH  Z 6Times New Roman Regular<09Z .Courier New Regular  i%$TT 4   7    ݀f XXThesalesbyrecipientsoutofincomeinkindmaybetootherhouseholds,butmayalsoinvolve  wholesaleandretailtradeasanactivity.Then,trademarginsshouldalsobeestimated.However,asthisisnotaspecificproblemforquarterlyaccounts,thiswillnotbefurtherdiscussedhere.#fX}X  l#(hH  Z 6Times New Roman Regular<9Z .Courier New Regular(;$( ` TT  j  ((3%$ ;!( ` TT       4   0   0@FF  1  ڀ(9 Z 6Times New Roman Regular(9 Z 6Times New Roman RegularWPCe; B81" B81""3'XXXX,,,,519"!hB8"`` ? 000GGGWWWgggwwwOg00OOggO'g0@OXgp0OgϘذOOgg0Og'O0g@OXgp0OgOg00OOggO'g0@OXgp0OgOOgg0Og'O0g@OXgp0OgOg00OOgg'O0g@OXgp0OgϘذOOgg0OgO'g0@OXgp0OgpH'W0X7`8g?h@pGxHψ_ؘp秀ǧϰ׿gggggggggpggggpgggggg6U}3M~%t~~ ~~ ~  ~  ]  [~<~ ~ ~  ~Q O M LKJI IH~qH H  B A @ ? > > = =  &>> > 9~p~cH`KH _KG  _L  h L HD L*B M|(C MP%  (D M"   *J N"   0M N#  2N N    3O O~]'MZ ,~\& Z Z4     3O O~  !Q !(6, 4N P      5 O P~Z  #+ R ~[ Y4$7- (    ` N Q )     _  O R~d1  `|~f5   E-, 1~p9* F +O- ~X ( *$    _  N T~^&!N' ~V    *_M80~z5Og~x 2}0~[8+G~V<  D }\^~XU+ +c--;*8             6                 5                                                                               ~j ! $U+ 124':~x*' T  V(; '8  +')~{@eY#$2'J, .)01 :$;~[L/ S-,F+5  =24<~Z2=O2c&>-/  2)4B 4~Y//"$##-- B A 0\%"$- *~W  )L--s  .+ 15~*Q-& K+> $ $# '$< A~W 1+nc ?G&L[$H +~]X,}$##4&p; . -= .:,<B~1n-E- Q$N &7< 6~XY*c28,%K%F,  # $ $. !S B~uQaH; EY#&-0R //,~ $n JQ- x " O#.  $E . 7~ &8! -K$F/:60:!X> F~v$A3x? 1   @1= .0 ! , -~2 %#P- U#%* '0-VG~w #5 [ !-U?(?kF / 8~#E;   .U . !0#.~j ,"K +G 7>'%".@6  JF~  'P"nsF>0.9 %~b.  A++UnH   %+=2,> +'~_2! yL O+=D9&0%!%~Z I#h (P "SE+  9'+$'(~V9 HVm?d$m J`# [2 &~W1  f M 2J?   (1 8'~V%1]){ \D$#7%8, 73%-#(~Z!2  <&  -2+,A)\>>&# $&*N1                                                                                   ~m *+ I   *(@(N @ -c%*[+                                                                         "                  %       (          i    rg  ~d~^ ^de!  ad  a  Q  b    b     b~g ,8~e     c  d~a    e !  e    e f  f  ggh# h"i!jj"k!l3l/m+n' o" p  r  tw P O O Q Q P P  N N~y"~z  Q S f~$~ #~~~ ,~  ~ ~~%t~~3!~3)WPC,,BB9B,, -B}?s? 3  ? ?? ?? ?> ? `?? 1|????C?ǀ??~???| 0tZZZ Z  6?  6  6  86?? ??7?? ???7p ??8?` 88?8 ?88?8? 8|p ?8` 8 ?   8    8?   7   87?0 ?  x7?0  ? x7?p  ? 7p> ? ? 7p<?? ? 7p0?  7 ? ? ?78?  ?7 ? ?' ?? ?' ? ? ??     x    ??? ???? ????? p? .????????? N?????? ???? N?<????????? N?@|~??| ???? ? N??>???? \????<<<~? ?\???xx ]|x?x` *?p.??? ?*??a.???>?y ??*???.????< 9  *??%??8 9 *??%?9   *?%?q?  *?~8?%8a?  ?*~p1?1 C? ? ?M??!! ? ? ?|??0#?? ?w ? ?p?0?? o ?  p?0? ?O~ ?  `00 ??? A` ?????? ` ? ?#?????    ?#????????`  #?????????  #?????/??? %??*??/>>??x %?*??/<|?~?? ??*??/|x?~? ??*???/x|?? ?? ?*?/p~??? ?*<??/a????? ???*?8???/???? ??^?0?????*???/ ?? ??*?p/ *??/????/???? ??"??? ? ??"???  ? ?"  ? ?  Y  x? x x x@??x?xy????x?|?x???x?xyyϼyy?y?y??yy?yyy??y?y????y?yǀzz z?z?zz??z{?  ??s? y σ<9 ;  | ||<<<VO $First Level, 12pt, bold, small caps, centered   @+  jV`$Second Level, 12pt, bold, centered  @+  VVV $Third Level, 12 pt, bold   fV)!$Fourth Level, 12pt, bold, italic  XV$Fifth Level, 12pt, italiczC4Tab ResetSets tabs back to default settings  X!/_3A<< cTTT6T*T5TBTKThT !( ` TT      4XXXX4SX4X T 7=/=dd7   9     A     :ECE:Workingpaperno.1  STATISTICALCOMMISSIONand A     COMMISSIONOFTHEEUROPEAN  ECONOMICCMMISSIONFOREUROPE A     COMMUNITIES(EUROSTAT) u CONFERENCEOFEUROPEANSTATISTICIANS  ORGANISATIONFORECONOMIC  @    9     A     COOPERATIONANDDEVELOPMENT JointUNECE/Eurostat/OECD ]   MeetingonNationalAccounts % `  (Paris,36June1997)Item1oftheprovisionalagenda }  ̛̜4 8 4S @NN-N#4S 4 8#@ NAgricultureinQuarterlyNationalAccounts:  @ NAllocationofOutputtoNonHarvestQuarters?44SԈ U #4S4#   SubmittedbyIMFL #  1         9 ,)5 98̛@NN-N̛8998̛̛̛̛̛899XT. I.  Introduction  I ̜  Therecordingofagriculturaloutputinthequarterlynationalaccountsposesspecificproblemsthatrelatetotheallocationofoutputtononharvestquarters.Thispaperwilldiscusstheseproblemsandvarioussolutions.ProblemsϜinthisrespectarecausedbythefactthattheoutputofacropmayonlyberealizedinadistinctharvestperiod,whilethegrowthofagriculturalproductsoftentakesplaceduringseveralaccountingperiods.Itisnotstraightforwardifthisoutputshouldonlybeattributedtotheharvestperiod,oralsotonon-harvestperiods.Ifoutputistobereallocatedtononharvestperiods,thequestionarisestowhatquartersandonwhatbasis.Furthermore,itshouldbeconsideredwhattheconsequenceswouldbeforothervariables.  Nogenerallyacceptedmethodologyisavailableconcerningthesereallocations.The1993SystemofNationalAccounts(SNA)statesthatinprinciple 9  cropsshouldbealsoallocatedtononharvestperiods,butisnotfullyconclusive.Furthermore,itleavesmanypracticalquestionsunanswered,andnointernationallyacceptedhandbookisyetavailablethataddressesthese.666P6 #  2       H    Onecouldarguethatquarterlyaccountsarenotveryusefulinagriculturaleconomies.Proponentsofthisviewwouldpointoutthatinsucheconomiesquarterlyaccountsdonotsupportbusinesscycleanalyses.Theywouldarguethatinthesecountriesshorttermeconomicgrowthasmeasuredthroughgrossdomesticproduct(GDP)isdominatedbyexogenousfactors(suchashoursofsunshineorinchesofrain)ratherthanbyunderlyingeconomicforces.However,otherusesofquarterlyaccountsdataaresufficientlyimportanttooutweighthisargument.First,animportantapplicationconcernstheuseofthestructuralinformationincludedinthequarterlyaccountsfortheformulationofeconomicandfinancialpolicies.Inthisrespect,itshouldbenotedthatamaturesetofquarterlyaccountswillincludeinformationconcerningexpendituresonGDP,abreakdownofvalueaddedbyindustryandbycomponents,andnationalincomeandsavingsdata.Second,quarterlyGDPcanbeusedasadenominatorforkeyindicatorssuchasthegovernmentdebt,thefiscaldeficit,andthequantityofmoney.̜  Furthermore,itcanbenotedthatquarterlynationalaccountsoftenservetoforecastannualdatasuchasGDPforacurrentyear,andprovideafirstestimateofthesedatabeforetheannualaccountsbecomeavailable.Also,quarterlynationalaccountsarevirtuallyindispensableinperiodsofrapidinflation,becauseunderthesecircumstancestheaxiomsunderpinningthenationalaccounts(suchaspricehomogeneitywithintheaccountingperiod)becomeuntenableonanannualbasis.̀  Thispaperwilltaketheusefulnessofquarterlyaccountsforgranted,eveninpredominantlyagriculturaleconomies,anditwillfocusonthespecificproblemsthatariseinthiscontext.FirstitwilldiscusstheprosandconsofϜreallocatingoutputtononharvestquartersandsubsequentlyalternativemethodstocarryoutreallocation.Subsequently,thepaperwilldiscussreallocationofϜotherquarterlyaccountsvariables.Finally,itwilldiscusssomepresentationdevicesthatmayhelpsomewhattomitigatetheproblemsusersmayhavewiththeuncertaintiesresultingfromreallocations. *H&1 Ї II.  Prosandconsofattributiontononharvestquarters  I   The1993SNAsupportstheallocationofagriculturaloutputtononharvest  quartersinprinciple,butrecognizesthepracticalproblemsinvolved.Inparagraph6.99itstates Inagriculture...itmaysometimesbenecessarytoestimatethevalueofworkinprogressinadvanceoftheproductionprocessbeingcompletedandthevalueofthefinishedproductsbeingknown.Italsostates,inparagraph6.100that Theremaybecircumstancesinwhichtheuncertaintiesattachedtotheestimationofthevalueofworkinprogressinadvanceoftheharvestaresogreatthatnousefulanalyticalorpolicypurposeisservedbycompilingsuchestimates.Forpracticalworkonthequarterlyaccounts,therefore,theformulationsdonotappearfullyconclusive,andafurtherexaminationoftheprosandconsofattributiontononharvestquartersseemsappropriate.Thisexaminationwillfocusontwoissues,viz.,(1)therecordingofproduction,and(2)whatconstitutesthe uncertaintiesthatparagraph6.100alludesto.  Toputthediscussionintoproperperspective,anobservationhastobemadeconcerningthetimingofthecompilationofthequarterlyaccounts.Oftenseveralversionsofquarterlyaccountarereleased,e.g.,anearlyversionshortlyaftertheendofthereferencequarter,afirstrevisiononequarterlater,andafinalversiontoachieveconsistencywiththeannualaccounts.Insomecountries,quarterlyaccountsareonlycompiledaftertheannualaccountshavebeenproduced,andthusfullyonanexpostbasis;thisissometimescalledquarterizationofthe A annualaccounts.However,themoregeneralcaseisthatquarterlyaccountsarecompiledduringtheongoingyear.Todistinguishthelattersituationfromquarterization,thispaperwillcallthesequarterlyaccounts earlyestimates. P Thecompilationofearlyestimatesclearlyhasadvantagesfromaninformationalpointofview,butposesproblemsthatanexpostapproachwouldnothave.This ) Мpaperwillfocusontheearlyestimates.   2.1  Therecordingofproduction  8 ̜  Concerningtherecordingofproduction,twoissueshaveparticularrelevanceinthiscontext,viz.,(1)timeofrecording,and(2)aggregation.Inrespectof  " timeofrecording ,thenationalaccountshaveafirmandlongstandingtraditionof X# accrualaccounting,implyingthattransactionshavetoberecordedatthetimetheytakeplace,andnotnecessarilyatthetimeoftheconcomitantmoneyflows.̜  Inrespectofproduction,accrualaccountinghasgivencausetothenotionofworkinprogress,anaccountingdeviceensuringattributionofoutputtotheperiodsinwhichproductionhasactuallytakenplace.Itispreciselythenotionofworkinprogressthatcausesthedilemmaoftheallocationofagriculturaloutput(crops)tononharvestquarters.  Thenotionofworkinprogressiswellestablished,notonlyinthenationalaccounts,butalsoinbusinessaccounting.Inthe1993SNA,therationalefor 9(#. workinprogresscanbefoundinthefollowingtwoquotations: Productioncanbebestdescribedingeneraltermsasanactivityinwhichanenterpriseproducesanoutput(paragraph6.6),and Forsimplicity,theoutputofmostgoodsorservicesisusuallyrecordedwhentheirproductionisfinished.However,whenittakesalongtimetoproduceaunitofoutput,itbecomesnecessarytorecognizethatoutputisbeingproducedcontinuously...(paragraph6.39.)Thus,therationaleforworkinprogressistheongoingproductionprocessinwhichinputsareused.  Thisrationaleseemsfullapplicabletoagriculture:longbeforeharvesttimeinputshavetobemade(fertilizer,seeds)andthereisanongoingproduction  0+8 processinvolvedinraisingthecrop(pestcontrol,irrigation,etc.)However,otherargumentscanbementionedthatpleadagainsttheapplicationoftheworkinprogressnotiontoagriculture.  Theseargumentsrelateinparticulartotheartificialityoftheconstruct.Tosomeextenttherecordingofworkinprogressdistortstheviewofincomegenerationandsavingbecauseoutputdoesnotgeneratemoneyflowsbeforeitissold.Thisseemsinparticularlyapplicabletoanonmonetizedorlargelynonmonetizedactivityasagriculture,especiallywhenitisonasubsistencebasisoronsmallfamilyscale.Inthesessituations,laborisoftenlargelysuppliedonafamilybasisandintermediateinputsmaybeselfproduced.Admittedly,thestandardnationalaccountsconceptsalsoapplytothesesituations.(Forinstance,onecouldarguethatasubsistencefarmerscroponthefieldishisworkinprogress,thatfromtheproductionofthisworkinprogresshederivesamixedincome,thathesavesallthisincomebecauseheislivingfromlastyearscrop,andthathedepletesthesesavingsatthetimeoftheactualharvest.)However,thiswouldseemtoofarfetchedtomany. 4S  #4S  *#JS4SRegarding aggregation ,twoissuesshouldbeconsidered,viz.,(1) JSԀ#JS  |+#MSJSaggregation Y OSMSofproducts,and(2)aggregationofunits.Concerningproducts,notethat 1 agriculturalproducersoftenengageinvariouscropswithdifferentgrowingandharvestingpatterns.Ifthisisthecase,theseasonalpatternoftotaloutputoftheseproducersmaybemuchlesspronounced#MSOS+#6 MS#MS  .#6thanoutputofindividualproducts. @   Concerningunits,anobviouspointworthemphasizingisthatnationalaccountsareprimarilyconcernedwithaggregates,andnotwithindividualunits.Animportantconsequenceinthiscontextisthatdifferencesbetweenindividualunitsconcerningthetimingofproductioncyclesdisappearatthelevelofanindustrygroupforthetotaleconomy.Forinstance,thismayoccurincountrieswithclimaticdifferencesbetweenregionsallowingharvestinginoneregionatthetimethatotherregionsengageinpreharvestactivities.Inthesecases,thespreadofcropsovertheyearismuchmorereflectiveofanongoingproductionprocess.̛ 2.2  Uncertainties  " ̜  Inmanyaspects,agricultureisamoreuncertainbusinessthanotherindustries,inparticularbecauseoftheeverpresentthreatofnaturaldisasterswreakinghavoconcropsonthefield.Althoughdisastersoccasionallyalsoaffectotherindustries,theygenerallydonothavethewidespreadeffectstheymayhaveinagriculture.Therefore,therecordingofdisastersinthenationalaccountsmeritsparticularexaminationforagriculture,andparticularlyforearlyquarterlyaccounts.̜  The1993SNAmakesadistinctionbetweenlossesfromdisastersand &p", catastrophiclosses.Disasterssuchasfiresareseenasnormalrisksofbusiness,andifinventoriesarelostasaresultofsuchphenomenathisistobetreatedaslossofoutput.Forinstance,ifafiredestroysinventoriesofworkinprogressinafurniturefactorytheoutputrelatedtotheseinventoriesiszero.However,ifaneventisdeemedacatastrophe,outputremainsunchangedandthecatastrophiclossofinventorieshastoberecordedintheotherchangesinvolumeofassetsϜaccounts.Thus,catastrophiclossesdonotaffectthevalueofproduction.  Appliedtoquarterlyagriculturaloutput,theserecommendationsimplythatwasteofcropsduetonormalevents(suchasinclementweather)aretoberecordedaslossofoutput,andthatlossofcropsduetocatastrophiclosseshavetoberecordedasotherchangesinvolumeofassets;theselosseswouldnotaffectthe 0+8 outputestimate.ThisappliesbothtotheoutputofharvestandnonharvestϜquarters:Ifthepotentialorestimateharvestoutputisattributedtononharvestquarterstheallocationshouldbeexclusiveofthewasteofcropsduetonormalevents,butinclusiveofcropslostthroughcatastrophes.  Fromthis,twouncertaintiesarise.First,inthecaseofnormallosseseitheranestimatehastobemadeoftheselossesbeforeallocatingoutputtononharvestquarters,orrevisionshavetobemadeafterthelossesbecomeknown.Second,thedistinctionbetweenanormaleventandacatastrophiclossissomewhatarbitrary,andthe1993SNAdoesnotprovidesolidgroundonwhichtobasea Q   distinction.8 #  3           2.3  Conclusion   `    Thespectrumofcountriesforwhichallocationofagriculturaloutputtononharvestquarterswouldbeusefulmaybelimitedbecause,atthelevelofthetotaleconomy,cropcyclesmaysmoothoutovertheyear.Also,theartificialityofrecordingcropsinthefieldasworkinprogressarguesagainstsuchallocation.Inadditiontothis,theuncertaintiesmentionedaboveconsiderablyreducethevalueofmakingsuchallocations.Furthermore,asthenextsectionswillshow,toachieveallocationfurtheruncertaintieshavetobeovercome.Allinall,thisleadstotheconclusionthatinmostcasesthequotationfromparagraph6.100ofthe1993SNAwouldapply ...nousefulpurposeisservedbycompiling y0 suchestimates(seepage2forthefullquotation.) III.  Alternativeallocationmethods     Ifitisdecidedthattheharvestoutputofaparticularcrophastobeallocatedtononharvestquarters,thenextquestiontobeansweredishowtoachieveallocation.Severalmethodscouldbedevelopedtoallocateaharvestoutputtopreharvestperiods;thispaperwilldiscussthree,viz.,(1)seasonaladjustmentmethods,(2)acostplusmarkupapproach,and(3)cropestimates.̜  Beforediscussingthesealternatives,abasicquestionhastobeanswered;viz.,towhatnonharvestperiodsoutputshouldbeallocated.The1993SNArecommendsthatallocationshouldbemadeto quartersinwhichproductiontakesplace.(Seeparagraph10.107.)Itmaynotalwaysbeselfevidenttowhatquartersthispertains.Forinstance,incountrieswhereallcultivationactivitiesceaseduringwinter,itcouldbearguedthatneverthelessproductionactivitiestakeplaceduringtheseperiods.Insupportofthis,itcouldbementionedthatthe "x' agriculturalenterprisesremaininoperation,andland,labor,andfixedcapitalassetsaretiedupeveniftheyarenotused.Furthermore,payments(suchasrent)maybemadeduringthisperiod.However,the1993SNAdefinesproductionas ...a  physicalprocess,...,inwhichlaborandassetsareusedtotransforminputsofgoodsandservicesintooutputsofothergoodsandservices.(Seeparagraph1.20).Thisimpliesthatinnongrowthquartersnoproductiontakesplace,andthatnoallocationshouldbemadetothesequarters.Apracticalconsiderationinsupportofthisisthatdecidingwhatoutputshouldbeallocatedwouldbedifficultifanoncultivationperiodisfollowedbyseveralcrops. 3.1  Seasonaladjustment     ̜  Proponentsofnonallocationmayarguethatthedistortionsthatnonallocationmaycausetoseasonalproductionpatternscanberemovedthroughex q(  postseasonaladjustments.However,asarguedbefore,expostquarterizationhas 9  lessvaluefromaninformationalpointofview.̜  Analternativewouldbetouseseasonaladjustmentmethods(suchastheCensusX12)toforecastoutputforperiodsbeforeaharvest.However,thiswouldimplyamodelingapproachthatisnotbasedonobserveddatafromtheperiodsunderconsideration.Also,itcouldbenotedthatmultiplicativeseasonaladjustmentmethodsdonotaffectperiodswithazerooutput(thusleavingnoncultivationperiodswithoutanattributedoutput),andthatadditivemethodshavenootherresultthanimputinganaverageoutput.Furthermore,itcanbenotedthatinviewofthemanyfundamentalandpracticalproblemswithseasonaladjustmentproceduresthiskindofdataisseenbymanyasanalyticaldata. 3.2  Estimatesbasedoncosts  a ̜  The1993SNArecommendsacostsplusmarkupapproachthatprovidesfirst  estimatesbasedoncosts,tobefollowedbyarevisionatthetimetheeventualvalueofthecropbecomesknown. #  4      כThecostsincludelabor,intermediateuse,and 8 fixedcapitalconsumption.Otheroutlaysconcerningproduction(suchasrents)canbeseenascoveredthroughthemarkup.̜  Consumptionoffixedcapitalcanbeseenastakingplaceonacontinuousbasis,nomattertheuseofthefixedcapitalassets.Thiswouldimplyallocationofoutputtoallquartersevenifnoproductiontakesplace.However,thiswouldnotagreewiththeaboveinterpretationofthe1993SNAsviewsonproductionand !& itisrecommendedheretorestricttheallocationoffixedcapitalconsumptiontoproductionquarters.̜  Itshouldbeputtothetestwhetheracostsplusmarkupapproachispracticableincountriesinwhichagriculturalactivitiesaremainlyfamilybased.Insuchasituation,agooddealoftheinputs(e.g.,labor,seedforsowing,fixedcapitalassets)isnotprovidedthroughthemarket,butwithinthehouseholdenterprise.Therelatedcostsarenotexpressedasamonetaryvalue,andthusseemaweakbasisforanoutputestimate.Furthermore,itcanbearguedthatinthesecircumstanceslaborinputsarelargelysupplydrivenanddonothaveastrongrelationtooutput.(Thisapplies,inparticular,tosmallscalefamilyfarming *H&1 that,forseveralreasons,isoftenoverlaborintensive.I #  5      )Onapracticalnoteit I canbeaddedthat,evenifdataareavailableontheacquisitionofintermediategoods,itseemsunlikelythatmuchisknownaboutthetimingoftheiruse. 3.3  Cropestimates  i  ̜  Athirdmethodforoutputattributionistofirstestimateacertaincropandsubsequentlyattributethisestimatetopreharvestquarters.Themaintwoproblemstobesolvedforthisapproachare(1)theestimateoftheeventualcropand(2)theattributionofthisoutputtothenonharvestquarters. 3.3.1  Theestimateoftheeventualcrop     ̜  Anestimateoftheeventualcropisneededbecause,obviously,duringpreharvestquartersnoobservedoutputdataareavailable.Onealternativefortheestimateoftheeventualcropwouldbetofirstestimatethevolumeofacrop,andsubsequentlyvaluethisvolumeatarelevantprice.Avolumeestimatecouldbebasedonanestimateoftheacreageundercultivationforaparticularcrop,combinedwithyieldestimates.Theacreageundercultivationestimatescouldbebasedonsurveys,oronaerialandsatellitephotography;yieldestimatescouldbebasedonaveragecropyields,andrevisedonthebasisofexpertviewsasneeded.Itmaybesurmisedthatinmanyagriculturalcountriesthiskindofinformationisavailable.̜5  5Concerningtherelevantprice,intheorytwoalternativescouldbeϜconsidered:(1)theaveragepriceofthecropduringtheperiodthatitismarketed,and(2)thepriceatthetimeoftheharvest.Theaveragepriceduringthetimeacropisonthemarketwillincludeaserviceelement(relatingtostorage)andaninflationelement,andisnotsuitableforthesereasons.Therefore,forthevaluationoftheeventualcrop,thepriceatthetimeoftheharvestseemsthemostappropriate(thisdoesnotmeanthatitshouldalsobeappliedtothepreharvestperiods;seebelow).̜  Alternativesfortheestimateofthepriceofthecropatharvesttimewouldbepricesonforwardmarkets,worldmarketprices,orthepreviouscropspricescorrectedforgeneralinflation.FuturemarketspricesmaynotalwaysbelocallyϜavailableandmayalsoturnouttobeoffthemark.Worldmarketpricesmaybelessrelevantforlessdevelopedmarketsandforproductsthatarepredominantlyconsumeddomestically.Therefore,amongthevariousalternatives,aharvestpriceestimatebasedonthepriceofthepreviouscropadjustedwithsomegeneralinflationindicatorseemsthemostappropriate.̜  Alternativesforsuchageneralinflationindicatorwouldincludetheconsumerpriceindex(CPI)andtheimplicitdeflatorofgrossdomesticexpenditures.Ineithercase,exclusionofpricesofimportedgoodsisrecommended.̜ 3.3.2  Theattributionoftheeventualcroptononharvestquarters  )$/ ̜  Threealternativesforaquarterlyattributionwouldbe(1)astraightlineinterpolation,(2)anattributionasafunctionofinputsovertime,and(3)anevenattributiontotherelevantquarters.Straightlineinterpolationhasan !,'3 intuitiveappeal,asitresemblesagrowthpatternovertime.However,thismethodwillprobablyentailnegativevalueaddedandincomeflowsinearlyperiods,whichseemscounterintuitive.Also,thereseemsnotheoreticalsoundreasontoattributelessoutputtoearlierperiods.̜  Amoresubtlemethodwouldbeanattributionasafunctionofinputsovertime.Thiswouldemphasizeproductionintheplantingandtheharvestseason,whichintuitivelyseemsrealistic.Aboveithasbeenarguedthatthecorrelationbetweeninputsandoutputsmaybeweak,butthisseemslessofaproblemfortheallocationofanindependentlyestimatedoutputthanfortheestimationoftheoutputitself.However,itmaybeproblematictocollectdataonallrelevantoutlays,whichshouldencompass(includingtransactionsinkind)labor,intermediateinputs,consumptionoffixedcapital,andrent.  Thethirdmethodtheevenattributionofproductiontoallrelevantquartersdoesnothaveanobviousrationaleeither,butdoesnotposeanyoftheproblemsmentionedabove.Becauseofthemanyuncertaintiessurroundingestimatesandbasicdata,thismaybethemoststraightforwardmethod.-  -Inapplyingthismethod,adistinctionshouldbemadebetweenconstantpricedataandcurrentpricedata.Forconstantpricedata,themoststraightforwardmethodwouldbetousethequantityestimateofthenewcropasallocatedtothequartersasavolumeindicator.Forthecurrentpriceestimatesofthepreharvestquarters,theestimatedpriceofthecropatharvesttimecannotbeuseddirectly,becausethiswouldincludeholdinggainsintheestimates.Asimplemethodtobypassthisproblemwouldbeastraightlineinterpolationbetweenthepreviouscropsharvestpriceandtheexpectedharvestpriceofthecropundercultivation.Iftheharvestpriceofthiscropistobeestimatedbasedonthelastcropspriceadjustedforgeneralinflation,anobviousrefinementwouldbetoapplythesamemethodforthequarterlyallocations.  Whateverthemethodofallocation,revisionswillbenecessaryasobserveddatabecomeavailableaftertheharvest.Concerningprices,itfollowsfromtheprecedingparagraphthattheearlierestimatescannotsimplybereplacedbytheharvestprice,becausethatwouldintroduceaninflationaryelementinthevaluationthatshouldbetreatedasholdinggains.Therefore,thepricesofthepreharvestallocationsshouldberevisedsogeneralinflationisexcluded(e.g.,throughastraightlineinterpolationbetweenlastandcurrentcropsprices). IV.  AllocationofOtherVariables  "x' ̜  Thusfar,thispaperhasfocussedontheallocationofoutputtononharvestquarters.However,becauseamaturesetofquarterlyaccountsalsowillcoverothervariables,theeffectsonthesevariablesshouldalsobeexamined.Thereisnouniversallyacceptedpracticetosuggestwhatvariablesquarterlyaccountsshouldcomprise.ThepositiontakenhereisthatamaturesetofquarterlyaccountswillincludeinformationconcerningexpendituresonGDP,abreakdownofvalueaddedbyindustryandbycomponents,andnationalincomeandsavingsdata.̜  TheSNAaccountsaffectedbyallocationofoutputtononharvestquartersincludetheGoodsandServicesAccounts,theProductionAccounts,theGenerationofIncomeAccount,theDistributionofIncomeAccounts,andtheUseofIncomeAccount.FromthevariablesoftheGoodsandServicesAccount,evidentlytotaloutputisaffectedbyanallocationofagriculturaloutput.Furthermore,changesininventoriesareaffected,becausethisisthecategoryunderwhichworkinprogressisclassifiedinthesystem.Taxesandsubsidiesonproductsareaffectedbytheallocationofoutputiftheyareonanaccrualbasis,becausethechanged  0+8 timingoftheoutputalsoimpactsontheperiodsoverwhichthetaxesandsubsidiesaccrue.Regardingintermediateuse,outputallocationdoesnothaveadirectimpact,becauseinprincipleinputsshouldberecordedinthequartersinwhichtheyareactuallymade.However,inpracticenotmuchinformationwillbeavailableonwhentheuseofintermediateinputsinagriculturetakesplace,andallocationofintermediateinputsconsistentwiththeoutputallocationmaybemostpractical.Fixedcapitalaccumulationisnotaffectedbecausecropoutputsdonotconcernfixedcapitalassets./+//]/ #  6        x М  ̜  Concerningtheallocationofconsumption,itcanalsobearguedthatitisnotaffectedbyanallocationofoutputtoquarters,becauseworkinprogressdoesnotbecomeavailableforuseoutsidetheproducingunit.However,allocationofconsumptionshouldbeconsideredforotherreasons.Inparticular,ifwagesandotherprimaryincomesarepaidinkindadecisionhastobemadeconcerningthetimeofrecording.̜.  .Especiallywithsubsistencefarming,wagesmaybepaidoutofthecropaftertheharvest.Thesameappliestorents,whichareoftenpaidonasharecroppingbasisinwhichthelandlordreceivespartofthecropaftertheharvest.Also,theoperatingsurplusofthefarmerandthedisposableincomethatresultsafterincomedistributiontransactionshavebeensettled,maybeinkindratherthaninmoney.̜  Thedelayedincomesassuchdonotposeaconceptualproblem,althoughcollectingtherelevantdatamaybedifficult.Conceptually,thedeferralofwagesandrentsgeneratesafinancialclaimofthebeneficiariesonthefarmerandaliabilityofthelatter.Theseclaimsandliabilitiesareredeemedatthetimeofpayment;viz.,atharvesttime.Inpractice,therecordingoftheseclaimsandliabilitiesatthetimetheyaregeneratedmaybehamperedbylackofdata,inwhichcaseanevendistributionoverrelevantnonharvestperiodsisanextbestsolution.̜  Amorefundamentalproblemconcernsconsumptionoutoftheseincomesinkind.Inparticular,thisaffectsincomeinkindoffoodcrops.Incomeinkindofnonfoodcropswillprobablybemonetizeduponreceipt,butfoodproductsarelikelytoberetainedfullyorpartlyforfinalconsumptionofthehouseholds.111e1 #  7       X# Thisposesaproblemforthetimeoftherecordingoffinalconsumptionbyhouseholds.̜  Concerningthetimeofrecordingofthisfinalconsumption,the1993SNA "x' statesthat Inpractice,theSystemmeasureshouseholdconsumptiononlyby #@( expendituresandacquisitions(seeparagraph9.40).Ifthiswouldbeappliedliterally,forsubsistencefarmingthiswouldmeanthatthebulkoffinalconsumptionfromthecropwouldbeconcentratedintheharvestquarters,whichdoesnotseemrealistic.Itshouldbenotedthattheabovequotationisfroma &`", 1993SNAparagraphdealingwithconsumptionofconsumerdurables,andthe I argumentsunderpinningthispracticalrecommendationdonotapplytosubsistencefarming.Forthisreason,andbecauseoftheunrealisticoutcomethatwouldresult,hereanotherapproachisrecommended.̜  Thisrecommendedapproachistoallocatethefinalhouseholdconsumptionexpenditureoutofincomesinkindtotheperiodsinwhichtheproductsareusedup.Thisimpliesthathouseholdsareseenasbuildingupastockofgoodsatthetimeoftheharvest,whichtheydepleteovertheperiodtothenextharvest.Innationalaccountsterminologythiswouldimplychangesininventories.(Notethatthesechangesininventorieswouldaffectpostharvestperiods,incontrasttotheallocationsdiscussedabove.)Again,ifdataonthespreadoftheconsumptionovertimearelacking,anevenspreadovertherelevantquartersseemsanacceptableϜnextbestsolution.̜  OntheProductionAccount,theallocationofoutput,intermediateuse,andtaxesandsubsidiesonproductsclearlyalsoaffectstheallocationofvalueaddedtoquarters.̜3  3TheGenerationofIncomeAccountwillobviouslyshowtheallocationoftaxesandsubsidiesonproductsascarriedoverfromtheProductionAccount.However,concerningthecompositionofvalueaddedthesituationislessclear.Here,twoalternativesarepossible,viz.,(1)toallocatethechangesinvalueaddedfullytooperatingsurplus/mixedincome,or(2)toallocatethechangesinvalueaddedbothtowagesandoperatingsurplus/mixedincome.Ifwagesareactuallypaidintheperiodthatlaborishired,noallocationneedstobemadetothiscategory,andtheeffectsofoutputallocationonvalueaddedwouldbefullyabsorbedbyoperatingsurplus/mixedincome.However,ifwagesarepaidonlyaftertheharvest,asmayhappeninsubsistencefarming,attributionofchangesinvalueaddedtomixedincome/operatingsurplusonlywouldseverelydistortthepictureoftheeconomy.Asarguedabove,thesewagesshouldbeallocatedtotheperiodinwhichthelaborhasbeenprovided,andthedeferredpaymentsshouldberecordedthroughfinancialtransactions.̜  ConcerningthePrimaryDistributionofIncomeAccount,itcanbenotedthattheeffectsonwages,taxesandsubsidiesonproducts,andoperatingsurplus/mixedincomewilldirectlycarryovertothisaccount.However,asquarterlyaccountsasdefinedheredonotconsidertransactionsbetweendomesticsectorsandbecauseexternaltransactionsarenotlikelytobeaffectedbyanallocationofagriculturaloutput,thisaccountdoesnotneedfurtherconsiderationinthisrespect.Thechangesinvalueaddedandtaxesonproductsthroughtheallocationtononharvestquartersdirectlyfindtheirwayintotheclosingbalanceofthisaccount,viz.,nationalincome.̛̜  TheDistributionofIncomeAccountswouldalsoonlyshowtheeffectsonthebalanceofthisaccountforthetotaleconomy,viz.,nationaldisposableincome.̜  ConcerningtheUseofIncomeAccounts,theoutcomewoulddependonthechoiceconcerningtheallocationofconsumption.Ifconsumptionisnottobeallocated,thechangesindisposableincomewouldbefullyabsorbedbynationalsaving.Ifconsumptionistobeallocated,asarguedforsubsistencefarming,bothconsumptionandsavingwouldbeaffected.̜4  9(#. 4 V.  Presentation -h)5 ̜   Onegeneraldisadvantageofallocationstononharvestquartersisthatthey A/*7 wouldintroduceextrauncertaintytotheaccounts;therefore,ifacountrychooses  0+8 toapplythis,specialcareinthepresentationofthedataisneededtomitigatetheproblemsusersmayhavewiththeuncertaintiesresultingfromreallocations.̜  Threesuggestionscanbemadeinthisrespect.First,itisrecommendedthatthemethodologybedocumentedcarefullyandthedocumentationbewidelyaccessible,sothatusersareabletoformtheiropinion.Althoughthiswillnotenhancethereliabilityofthefigures,itwillatleastenableaviewonwhethertheyaresufficientlyreliableforvariouspurposes.Second,toserveuserswhodeemtheallocationsinsufficientlyreliableordonotcareforallocationsanyway,theseallocationsshouldbespecifiedandquantified.Third,thedatacouldbepresentedinclusiveandexclusiveoftheallocationsor,ifusersprefer,ofagriculturealtogether.